Applying Barlow Twins with sequence-tailored augmentations to user histories yields representations that outperform a dual-encoder baseline in several recommendation tasks with limited labels.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IR 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Enhancing User Sequence Modeling through Barlow Twins-based Self-Supervised Learning
Applying Barlow Twins with sequence-tailored augmentations to user histories yields representations that outperform a dual-encoder baseline in several recommendation tasks with limited labels.